U.S. patent number 6,861,617 [Application Number 10/453,732] was granted by the patent office on 2005-03-01 for method of reducing distortion by transient thermal tensioning.
This patent grant is currently assigned to Edison Welding Institute, Inc.. Invention is credited to Randal Martin Dull, James R. Dydo, James Joseph Russell, Janak Shanghvi.
United States Patent |
6,861,617 |
Dull , et al. |
March 1, 2005 |
Method of reducing distortion by transient thermal tensioning
Abstract
A method to minimize the distortion, such as buckling, caused in
the welding of thin plate by utilizing transient thermal tensioning
to induce areas of residual tensile stress. Distortion of such
welded plates after stiffeners are welded to the plates is due to
large areas of unsupported residual compressive stress following
the tensile stress that is induced along the weld lines by the
welding of the stiffeners. Application of transient thermal
tensioning by moving heat sources at the time of welding of the
stiffeners induces areas of residual tensile stress that minimize
the tendency of the plates to buckle. Multiple heat sources of
variable intensity may be utilized, and multiple stiffeners may be
welded to a single panel by the method. Minimization of distortion
improves the function and appearance of the finished plates and
minimizes post-welding repairs.
Inventors: |
Dull; Randal Martin (Columbus,
OH), Dydo; James R. (Groveport, OH), Russell; James
Joseph (Hilliard, OH), Shanghvi; Janak (Columbus,
OH) |
Assignee: |
Edison Welding Institute, Inc.
(Columbus, OH)
|
Family
ID: |
33489602 |
Appl.
No.: |
10/453,732 |
Filed: |
June 3, 2003 |
Current U.S.
Class: |
219/137R;
228/230 |
Current CPC
Class: |
B23K
9/32 (20130101); B23K 9/0256 (20130101) |
Current International
Class: |
B23K
9/025 (20060101); B23K 9/32 (20060101); B23K
009/00 () |
Field of
Search: |
;219/137R,136
;228/230,231 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
|
|
|
|
|
|
|
04-52079 |
|
Feb 1992 |
|
JP |
|
04-220176 |
|
Aug 1992 |
|
JP |
|
WO88/06505 |
|
Jul 1988 |
|
WO |
|
Other References
Guan, et al. "Low stress non-distortion welding" Welding in the
World 33:3,pp 160-167. .
Burak, et al. "Controlling the Longitudinal Plastic Shrinkage" Avt.
Svarka, 1977, 3:27-29. .
Michaleris, et al "Minimization of Welding Residual Stress" Welding
Res. Supp., 11/99, pp. 361-366. .
Michaleris and Sun, "Finite Element Analysis" Wel;ding Research
Supp. 11/97, pp. 451-457..
|
Primary Examiner: Shaw; Clifford C.
Attorney, Agent or Firm: Gallagher & Dawsey Co. L.P.
Gallagher; Michael J. Dawsey; David J.
Government Interests
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
This invention was made with government support under Contract No.
N00140-96-C-0188, Navy Joining Center Project No. 42372GDE awarded
by the Office of Naval Research (ONR). The government has certain
rights in the invention.
Claims
We claim:
1. A method of transient thermal tensioning at least one plate
during joining of at least one stiffener to the at least one plate
resulting in a stiffened plate, by welding to minimize distortion,
comprising the steps of: determining at least one location on the
at least one plate to apply at least one heat source, at a
predetermined lateral distance from the at least one stiffener and
at a predetermined separation distance from a surface of the at
least one plate, to minimize the propensity of the at least one
plate to distort by minimizing the propensity of the at least one
plate to buckle by altering the residual stress pattern in the at
least one plate; placing the at least one stiffener on the at least
one plate; locating at least one welding device in close proximity
to the at least one stiffener and the at least one plate; applying
the at least one heat source to the at least one location on the at
least one plate; energizing and moving the at least one welding
device to weld the at least one stiffener to the at least one
plate; and moving the at least one heat source in conjunction with
the travel of the at least one welding device.
2. The method of claim 1, wherein the step of determining the at
least one location on the at least one plate to apply the at least
one heat source to minimize the propensity of the at least one
plate to distort further includes minimization of the propensity of
the at least one plate to buckle.
3. The method of claim 2, wherein the step of determining the at
least one location on the at least one plate to apply the at least
one heat source to minimize the propensity of the at least one
plate to buckle further comprises the step of altering the residual
stress pattern in the at least one plate.
4. The method of claim 3, wherein the step of altering the residual
stress pattern in the at least one plate produces a location of the
at least one heat source resulting in a minimum buckling analysis
eigenvalue of the stiffened plate of at least 0.4.
5. The method of claim 1, wherein the step of altering the residual
stress pattern in the at least one plate further comprises the step
of disrupting at least one compressive stress pattern in the at
least one plate by inducing at least one area of tensile
stress.
6. The method of claim 1, wherein the step of determining the at
least one location on the at least one plate to apply the at least
one heat source to minimize the propensity of the at least one
plate to distort further includes locating a first location about a
first side of one of the at least one stiffener, at a first
predetermined lateral distance from the at least one stiffener, and
a second location about a second side of one of the at least one
stiffener, opposing the first side, at a second predetermined
lateral distance from the at least one stiffener.
7. The method of claim 1, wherein the at least one welding device
is a first welding device and a second welding device wherein the
first welding device and the second welding device concurrently
weld opposing sides of the at least one stiffener to the at least
one plate.
8. The method of claim 1, wherein the at least one welding device
is at least one arc.
9. The method of claim 1, wherein the at least one heat source is a
flame heater.
10. The method of claim 1, wherein the at least one heat source
further includes a plurality of heat sources being located at
different predetermined lateral distances from the at least one
stiffener.
11. The method of claim 1, wherein the at least one heat source
further includes a plurality of heat sources of variable heating
intensity.
12. A method of transient thermal tensioning at least one plate
during joining of at least one stiffener to the at least one plate
resulting in a stiffened plate, by welding to minimize distortion,
comprising the steps of: determining at least one location on the
at least one plate to apply at least one heat source, at a
predetermined lateral distance from the at least one stiffener and
at a predetermined separation distance from a surface of the at
least one plate, to minimize the propensity of the at least one
plate to buckle by altering the residual stress pattern in the at
least one plate by disrupting at least one compressive stress
pattern in the at least one plate by inducing at least one area of
tensile stress; placing the at least one stiffener on the at least
one plate; locating at least one welding device in close proximity
to the at least one stiffener and the at least one plate; applying
the at least one heat source to the at least one location on the at
least one plate; energizing and moving the at least one welding
device to weld the at least one stiffener to the at least one
plate; and moving the at least one heat source in conjunction with
the travel of the at least one welding device.
13. The method of claim 12, wherein the step of altering the
residual stress pattern in the at least one plate further comprises
the step of disrupting at least one compressive stress pattern in
the at least one plate by inducing at least one area of tensile
stress.
14. The method of claim 12, wherein the step of determining the at
least one location on the at least one plate to apply the at least
one heat source to minimize the propensity of the at least one
plate to buckle further includes locating a first location about a
first side of one of the at least one stiffener, at a first
predetermined lateral distance from the at least one stiffener, and
a second location about a second side of one of the at least one
stiffener, opposing the first side, at a second predetermined
lateral distance from the at least one stiffener.
15. The method of claim 12, wherein the at least one welding device
is a first welding device and a second welding device wherein the
first welding device and the second welding device concurrently
weld opposing sides of the at least one stiffener to the at least
one plate.
16. The method of claim 12, wherein the at least one welding device
is at least one arc.
17. The method of claim 12, wherein the at least one heat source is
a flame heater.
18. The method of claim 12, wherein the step of altering the
residual stress pattern in the at least one plate produces a
location of the at least one heat source resulting in a minimum
buckling analysis eigenvalue of the stiffened plate of at least
0.4.
19. The method of claim 12, wherein the at least one heat source
further includes a plurality of heat sources being located at
different predetermined lateral distances from the at least one
stiffener.
20. The method of claim 12, wherein the at least one heat source
further includes a plurality of heat sources of variable heating
intensity.
21. A method of transient thermal tensioning at least one plate
during joining of at least one stiffener to the at least one plate
resulting in a stiffened plate, by welding to minimize distortion,
comprising the steps of: determining by numerical modeling of the
stiffened plate and the process used to create the stiffened plate,
at least one location on the at least one plate to apply at least
one heat source, at a predetermined lateral distance from the at
least one stiffener and at a predetermined separation distance from
a surface of the at least one plate, to minimize the propensity of
the at least one plate to buckle by altering the residual stress
pattern in the at least one stiffened plate resulting in a minimum
buckling analysis eigenvalue of the stiffened plate of at least
0.4; placing the at least one stiffener on the at least one plate;
locating at least one welding device in close proximity to the at
least one stiffener and the at least one plate; applying the at
least one heat source to the at least one location on the at least
one plate; energizing and moving the at least one welding device to
weld the at least one stiffener to the at least one plate; and
moving the at least one heat source in conjunction with the travel
of the at least one welding device.
22. The method of claim 21, wherein the at least one heat source
further includes source a plurality of heat sources of variable
heating intensity being located at different predetermined lateral
distances from the at least one stiffener.
Description
TECHNICAL FIELD
The present invention relates to the field of material joining;
particularly, to a method of transient thermal tensioning of panels
during welding to reduce distortion.
BACKGROUND OF THE INVENTION
Numerous industries have long sought to reduce, or eliminate,
distortion of panels that have been joined by welding. Weld induced
distortion has plagued many industries, including the automobile,
aviation, heavy manufacturing, and shipbuilding industries, among
others, since the advent of welding.
Such distortion has become an increasingly common problem as
manufacturers increase the use of relatively thin plates of
material to construct various articles of manufacture in an effort
to reduce weight. The thin plates generally must be reinforced with
stiffeners welded to the plates to obtain the strength required for
a particular application. This stiffener welding introduces
residual stresses in the structure and may lead to several modes of
distortion, namely, transverse shrinkage, angular change,
rotational distortion, longitudinal shrinkage, longitudinal
bending, and buckling distortion. Hereinafter usage of the term
distortion shall refer to all types of distortion generally, unless
indicated otherwise. Thin plates are particularly susceptible to
buckling distortion due to their low bending stiffness compared to
their membrane stiffness. Buckling distortion is characterized by a
wavy undulating surface of the plate. Such out-of-plane distortion
is often many orders of magnitude greater than the thickness of the
plate and generally leads to the loss of dimensional control and
structural integrity.
During welding, a weld region's high temperature causes compressive
stress as a result of thermal expansion in the region and the
corresponding restraint on the expansion by the surrounding cooler
material. The compressive stress in the weld region may then exceed
the yield stress of the plate at the elevated temperature. As such,
material in the vicinity of the weld plastifies and compressive
plastic strains are produced. As the weld cools the stress patterns
change from compressive to tensile in the locations that have
plastified during the welding, thereby producing residual tensile
stress in the weld region.
Buckling distortion occurs if the residual compressive stress in
the plate exceeds the critical buckling stress of the assembly.
Therefore, stated another way, buckling distortion is a result of
the creation of residual tensile stress in the plate along each
stiffener and the residual compressive stress in the plate between
each stiffener and along any free edge of the plate.
Industry has tried to overcome buckling distortion in a number of
ways, both mechanically and thermally. Some industries, including
shipbuilding, have simply learned to accept buckling distortion and
apply a post-welding procedure to remedy the distortion. The
post-welding procedure is often referred to as "flame
straightening" and involves the heating of discrete spots of the
plate until they are red hot and then quenching the spots with
water to reduce the wavy nature of the stiffened plates. Shipyards
generally employ an entirely separate class of skilled tradespeople
known as flame straighteners to perform this function. Such flame
straightening is a trial and error approach that requires
tremendous skill and is extremely time consuming. Flame
straightening often requires the repainting of flame damaged areas.
In fact, studies have indicated that $3.4 million is spent
correcting distortion during the construction of each destroyer
built for the United States Navy.
Some have tried to overcome buckling distortion using mechanical
methods. An example of this has been called "low stress
non-distortion (LSND) welding" as reported by Guan, et al., in
their paper "Low stress non-distortion (LSND) welding--a new
technique for thin materials;" Welding in the World, 33:3, pp.
160-167 (1994). These methods are often referred to as "back
bending" and generally include some form of mechanical tensioning.
In the method of Guan, et al., a stretching effect is produced by
specific temperature distribution while restraining fixtures (e.g.,
"two-point:clamping") are used to prevent transient out-of-plane
buckling movement of the workpieces. As one with skill in the an
can imagine, mechanical tensioning of large plates is impractical
for most applications.
The most promising method for overcoming buckling distortion of
thin plates is known as thermal tensioning. Thermal tensioning is
characterized by the application of auxiliary heat during the
welding process. Thermal tensioning is divided into static thermal
tensioning and transient, or dynamic, thermal tensioning. Static
thermal tensioning is a technique for controlling welding residual
stress and distortion by generating tensile stress at the weld zone
prior to, and during welding, by imposing a predetermined steady
state temperature gradient. Achieving a predetermined steady state
temperature gradient requires the use of a combination of heating
elements and cooling elements to create a heat sink and achieve the
temperature gradient. Heating elements, often in the form of direct
fired heaters or resistive heating blankets, are applied on
opposing sides of the stiffener location at a predetermined
distance away from the stiffener. Cooling is then provided in the
immediate vicinity of the proposed weld location and is generally
accomplished with the impingement of cool water to the underside of
the plate. An exemplar of this technique is seen in the work of
Burak, et al, reported in "Controlling the Longitudinal Plastic
Shrinkage of Metal During Welding;" Avt. Svarka, No. 3, pp. 27-29
(1977).
Burak, et al., utilized electrical strip heating elements beneath
the plate and lateral to the weld line, with a water cooled copper
plate below the weld line, to produce the required gradient.
Welding the stiffener to the plate takes place once the desired
temperature differential is achieved. While carefully controlled
small scale laboratory experimentation have shown that static
thermal tensioning does reduce the amount of buckling distortion,
it is widely accepted that the use of steady state heating and
cooling would not be practical in a manufacturing environment due
to the time required to reach steady state and the limitations
associated with cooling elements.
Transient thermal tensioning utilizes a transient temperature
differential generally produced by two heating bands traveling
along with the welding torches that are joining the stiffener to
the plate. A large amount of research has been performed on the
transient thermal tensioning technique, as applied to the reduction
of residual stress. This research has primarily focused on
determining the appropriate intensity, size, and location of the
heat source to minimize welding residual stress, thereby reducing
the amount of distortion. More specifically, and most importantly,
the research has been directed to reducing the maximum value of the
peak stress, i.e. the maximum tensile stress, observed at the
stiffeners. A detailed analysis of thermal tensioning to minimize
tensile stress is seen in Michaleris and Sun, "Finite Element
Analysis of Thermal Tensioning Techniques Mitigating Weld Buckling
Distortion;" Welding Research Supplement, November 1997, pp. 451-s
thorough 457-s; and Michaleris, et al., "Minimization of Welding
Residual Stress and Distortion in Large Structures;" Welding
Research Supplement, November 1999, pp. 361-s through 366-s. This
focus on reducing the maximum value of the peak stress has not
provided the results necessary to make transient thermal tensioning
commercially viable.
In contrast, the method of the present invention makes no attempt
to reduce the peak stress, but rather focuses on altering the
stress pattern. By inducing areas of tensile stress in desirable
locations, the present invention achieves the primary goal, that of
greatly reducing the propensity of the plate to buckle.
Accordingly, the art has needed a means for minimizing distortion
that occurs concurrently with the welding of the stiffeners. While
some of the prior art devices attempted to improve the state of the
art, none have recognized the importance of achieving a desirable
stress pattern to reducing the propensity to buckle. Additionally,
the prior art has not been suitable for widespread application in a
manufacturing environment. The prior art has failed to achieve the
unique and novel configurations and capabilities of the present
invention. With these capabilities taken into consideration, the
instant invention addresses many of the shortcomings of the prior
art and offers significant benefits heretofore unavailable.
Further, none of the above inventions and patents, taken either
singly or in combination, is seen to describe the instant invention
as claimed.
SUMMARY OF INVENTION
In its most general configuration, the present invention advances
the state of the art with a variety of new capabilities and
overcomes many of the shortcomings of prior methods in new and
novel ways. In its most general sense, the present invention
overcomes the shortcomings and limitations of the prior art in any
of a number of generally effective configurations. An object of the
invention is to minimize buckling resulting from the welding of
stiffeners to relatively thin plates by utilizing transient thermal
tensioning to create areas of tensile stress in the welded plate
that interrupt areas of compressive stress. The instant invention
demonstrates such capabilities and overcomes many of the
shortcomings of prior methods in new and novel ways. While
disclosed primarily in the welding of stiffeners to plate, the
transient thermal tensioning of the instant invention may be used
in many other welding applications, such as, by way of example and
not limitation, the butt welding of sheets.
In one of the many preferable configurations, the method comprises
a method of transient thermal tensioning that creates areas of
tensile stress that interrupt areas of residual compressive stress
at desirable locations, and thereby reduce the tendency of welded
plate to buckle. The method comprises, in general, the steps of
applying transient thermal tensioning to reduce buckling distortion
involved by applying a carefully controlled amount of heat in a
specific pattern at a specific distance from a weld, and moving the
area of heat along with the welding torch as the weld is made. The
distance of the heat from the weld varies with the characteristics
of the weldments.
These variations, modifications, alternatives, and alterations of
the various preferred embodiments, processes, and methods may be
used alone or in combination with one another as will become more
readily apparent to those with skill in the art with reference to
the following detailed description of the preferred embodiments and
the accompanying figures and drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
Without limiting the scope of the present invention as claimed
below and referring now to the drawings and figures:
FIG. 1 shows, in transverse section, a residual stress profile of
an experimental plate with a single welded longitudinal
stiffener;
FIG. 2 shows, in transverse section, a residual stress profile of
an experimental plate with a single welded longitudinal stiffener
that has been welded utilizing the method of transient thermal
tensioning of the instant invention;
FIG. 3 shows an elevated perspective view, not to scale, of the
plate, stiffener, welding means, and heat sources of one embodiment
of the instant invention;
FIG. 4 shows an elevated perspective view, not to scale, of a plate
having three longitudinal stiffeners welded approximately
equidistantly along the long axis of the plate;
FIG. 5 shows an elevated perspective view, not to scale, showing an
8 foot by 20 foot experimental plate with a first stiffener placed
approximately two feet from one edge of the plate, in welding
position, with one heat source placed approximately twenty-one
inches from the centerline of the stiffener;
FIG. 6 shows an elevated perspective view, not to scale, showing an
8 foot by 20 foot experimental plate with a first stiffener welded
to the plate, a second stiffener in welding position, and two heat
sources, each heat source placed approximately forty-four inches
from the centerline of the second stiffener;
FIG. 7 shows an elevated perspective view, not to scale, showing an
8 foot by 20 foot experimental plate with a first and second
stiffener welded to the plate, a third stiffener in welding
position, and one heat source placed approximately twenty-one
inches from the centerline of the third stiffener;
FIG. 8 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 1/2 inch from a plate edge,
along the longitudinal length (horizontal axis), of an 8 foot by 20
foot experimental plate reinforced with three welded longitudinal
stiffeners, having been welded according to the method of the
instant invention utilizing transient thermal tensioning;
FIG. 9 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 1/2 inch from a plate edge,
along the longitudinal length (horizontal axis), of an 8 foot by 20
foot experimental plate reinforced with three welded longitudinal
stiffeners, having been welded without utilizing transient thermal
tensioning;
FIG. 10 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 361/2 inches from a plate
edge, along the longitudinal length (horizontal axis), of an 8 foot
by 20 foot experimental plate reinforced with three welded
longitudinal stiffeners, having to been welded according to the
method of the instant invention utilizing transient thermal
tensioning;
FIG. 11 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 361/2 inches from a plate
edge, along the longitudinal length (horizontal axis), of an 8 foot
by 20 foot experimental plate reinforced with three welded
longitudinal stiffeners, having been welded without utilizing
transient thermal tensioning;
FIG. 12 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 601/2 inches from a plate
edge, along the longitudinal length (horizontal axis), of an 8 foot
by 20 foot experimental plate reinforced with three welded
longitudinal stiffeners, having been welded according to the method
of the instant invention utilizing transient thermal
tensioning;
FIG. 13 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 601/2 inches from a plate
edge, along the longitudinal length (horizontal axis), of an 8 foot
by 20 foot experimental plate reinforced with three welded
longitudinal stiffeners, having been without utilizing transient
thermal tensioning;
FIG. 14 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 951/2 inch from a plate
edge, along the longitudinal length (horizontal axis), of an 8 foot
by 20 foot experimental plate reinforced with three welded
longitudinal stiffeners, having been welded according to the method
of the instant invention utilizing transient thermal
tensioning;
FIG. 15 shows a graph of the measured amount of out-of-plane
distortion (vertical axis) measured at 951/2 inch from a plate
edge, along the longitudinal length (horizontal axis), of an 8 foot
by 20 foot experimental plate reinforced with three welded
longitudinal stiffeners, having been welded without utilizing
transient thermal tensioning; and
FIG. 16 shows a graph of the residual longitudinal stress profile,
measured in transverse section, of an 8 foot by 20 foot
experimental plate after a first, (top; labeled Pass One), (middle,
labeled Pass Two), and third (bottom; labeled Pass Three,
longitudinal stiffener or stiffeners has been welded, as
illustrated in FIG. 5, according to the method of the instant
invention utilizing transient thermal tensioning.
DETAILED DESCRIPTION OF THE INVENTION
The method of transient thermal tensioning of the present invention
enables a significant advance in the state of the art. The
preferred embodiments of the method accomplish this by new and
novel arrangements of elements and methods that are configured in
unique and novel ways and which demonstrate previously unavailable
but preferred and desirable capabilities. In particular, the method
is a low-cost, easily reproducible technique that is attractive in
the manufacturing environment because it is simple to apply,
requires virtually no equipment modification, and is
environmentally friendly. The detailed description set forth below
in connection with the drawings is intended merely as a description
of the presently preferred embodiments of the invention, and is not
intended to represent the only form in which the present invention
may be constructed or utilized. The description sets forth the
designs, functions, means, and methods of implementing the
invention in connection with the illustrated embodiments. It is to
be understood, however, that the same or equivalent functions and
features may be accomplished by different embodiments that are also
intended to be encompassed within the spirit and scope of the
invention.
In order to reduce weight, many industries are increasingly
favoring the use of stiffened thin plates to construct various
assemblies. The plates are stiffened by welding stiffeners on to
them. The welding procedure introduces residual stresses in the
structure. This leads to several modes of distortion, namely
transverse shrinkage, angular change, rotational distortion,
longitudinal shrinkage, longitudinal bending, and buckling
distortion. Thin plates have a low is bending stiffness compared to
their membrane stiffness. This makes them particularly susceptible
to buckling. In fact, the dominant mode of distortion induced by
welding in these structures is buckling of the plate. Of the six
modes of distortion, buckling distortion is the only one that is
not linearly proportional to the stress pattern. In fact, buckling
is an inherently unstable phenomenon and can therefore often be
mitigated by changes in the stress pattern. In other words, as weld
induced residual stress is applied to the plate a proportional
amount of distortion is caused. As the residual stress increases,
the distortion increases up to the instability point and then
unstable buckling occurs.
Buckling is caused by compressive stresses in the plate. The
compressive stress in the plate arises to counter the tensile
stress induced at the weld. Generally, the areas of the plate that
are subject to a heating process, either by welding or application
of an auxiliary heat source, remain in tension (positive residual
stress) when the plate cools to room temperature. Alternatively,
areas of the plate that are adjacent to the heated areas, end up in
compression (negative residual stress) when the plate cools to room
temperature. Large expanses of the plate, between stiffeners, are
typically in compression after welding. It is these unsupported
areas between the stiffeners that induce the observed buckling.
While disclosed primarily in the welding of stiffeners to plate,
the transient thermal tensioning of the instant invention may be
used in many other welding applications, such as, by way of example
and not limitation, the butt welding of sheets.
An example of a single stiffener welded to a relatively thin plate
is illustrated in FIG. 1. Taken in transverse section, a single
stiffener 100 is shown welded in a longitudinal direction at
approximately the center of the plate 200. The residual stress
profile at various points along the transverse section is shown.
Areas above the horizontal line, or zero stress point, represent
areas that are in residual tensile stress. Areas below the
horizontal line, or zero stress point, represent areas that are in
residual compressive stress. It can be seen that there exists an
area of maximal tensile stress at the weld lines of the stiffener
100, and that stress drops to compression slightly lateral to the
stiffener 100, and remains in slightly decreasing compression to
the edge of the plate 200. To maintain the plate 200 in
equilibrium, the integral of the residual stress, taken over the
width of the plate 200, must sum to zero.
Traditional efforts to reduce buckling center around reducing the
tensile residual stress induced during the welding process. Such
traditional techniques include pre-weld chilling, static thermal
tensioning, and many others. All these processes have associated
costs and need to be reconciled to the manufacturing set-up already
in place.
The method of transient thermal tensioning of the present invention
focuses on reducing a stiffened plate's propensity to distort when
joining, by welding, at least one plate 200 and at least one
stiffener 100. The method of transient thermal tensioning of the
present invention consists generally of placing the at least one
stiffener 100 on the at least one plate 200; locating at least one
welding device 300 in close proximity to the at least one stiffener
100 and the at least one plate 200; determining at least one
location 412 on the at least one plate 200 to apply at least one
heat source 400, at a predetermined lateral distance 414 from the
at least one stiffener 100 and at a predetermined separation
distance 430 from a surface 210 of the at least one plate 200, to
minimize the propensity of the at least one plate 200 to distort;
applying the at least one heat source 400 to the at least one
location on the at least one plate 200; energizing and moving the
at least one welding device 300 to weld the at least one stiffener
100 to the at least one plate 200; and moving the at least one heat
source 400 in conjunction with the travel of the at least one
welding device 300. While the instant invention is illustrated
herein as a single or paired welding device 300 operating with a
single or paired heat source 400, the use of multiple welding
devices 300 and multiple heat sources 400 is contemplated. In
particular, the method may be used to weld multiple stiffeners 100
at the same time. Additionally, the method does not contemplate a
particular order of welding stiffeners 100 to plates 200, that is,
stiffeners 100may be welded beginning near one lateral edge of the
plate 200 and then progressing in a predetermined pattern across
the plate 200, or a central stiffener 100 may be welded near the
center of the plate 200 and subsequent stiffeners 100 welded
laterally to the central stiffener 100, or some other pattern of
stiffener welding may be optimized to the individual demands of a
particular situation.
The method of the present invention takes the approach of altering
the residual stress in the at least one plate 200 rather than
minimizing or eliminating it, as with prior art attempts at
reducing distortion. For example, a single stiffener 100 welded to
a single plate 200 with residual stress altered by transient
thermal tensioning is illustrated in FIG. 2. Taken in transverse
section, a single stiffener 100 is shown welded in a longitudinal
direction at approximately the center of the plate 200. Transient
thermal tensioning according to the method of the instant invention
has been applied by heat sources 400, lateral to the single
stiffener 100, as also seen in FIG. 2, during the welding process
as described herein. Shown below the transverse section is the
residual stress profile along the transverse section. Areas above
the horizontal line, or zero stress point, represent areas that are
in residual tensile stress. Areas below the horizontal line, or
zero stress point, represent areas that are in residual compressive
stress. It can be seen that there exists an area of maximal tensile
stress at the weld lines of the stiffener 100, and that stress
drops into compression moving lateral to the stiffener 100. Near
the edge of the plate 200, the areas exposed to transient thermal
tensioning exist in tensile stress. To maintain the plate 200 in
equilibrium, the integral of the residual stress, taken over the
width of the plate 200, must sum to zero.
Because the areas of transient thermal tensioning induced tensile
stress tend to increase the strain energy required for out of plane
bending, a buckling analysis of the plate of FIG. 2 has a higher
eigenvalue than that of the plate of FIG. 1, and is less likely to
buckle. If the calculation of eigenvalues produces any that are
less than 1, the plate is predicted to buckle. Experimentally, the
plate 200 as it is shown welded in FIG. 1, has a minimum cigenvalue
of 0.58, and would therefore be expected to buckle. The plate 200
as it is shown in FIG. 2, welded with the application of transient
thermal tensioning according to the instant invention, has a
calculated minimum eigenvalue of 2.96, and therefore would not be
expected to buckle. In fact, the relative propensity of the plates
of FIGS. 1 and 2 to buckle maybe expressed as the quotient of their
calculated minimum eigenvalues, that is, the plate as welded in
FIG. 2, with the application of transient thermal tensioning
according to the instant invention, has over five times the
expected resistance to buckling of the plate of FIG. 1
(2.96/0.58=5.1034)
The step of determining the at least one location 412 on the at
least one plate 200 to apply the at least one heat source 400 to
minimize the propensity of the at least one plate 200 to distort
may further include the minimization of the propensity of the at
least one plate 200 to buckle. The propensity to buckle depends on
a large number of factors. For rectangular unstiffened plates, the
propensity to buckle may be calculated using closed-form equations
dependent upon the plate 200 thickness, the ratio of the plate 200
width to the length, and the elastic modulus of the material. When
stiffeners 100 are added to a plate 200, additional variables enter
into the calculation, including the geometry of the stiffeners 100
and the spacing of the stiffeners 100. When stiffeners 100 are
welded to a plate 200, closed-form solutions are inadequate to
predict the propensity to buckle and numerical modeling is
required.
Still further, the step of determining the at least one location
412 on the at least one plate 200 to apply the at least one heat
source 400 to minimize the propensity of the at least one plate 200
to buckle may also further comprise the step of altering the
residual stress in the at least one plate 200. By altering the
residual stress in the at least one plate 200 the stress-stiffness
of the structure is enhanced and the propensity to buckle is
reduced. Because the out-of-plane distortion is not linearly
proportional to the stress, small changes in stress result in
dramatic reductions in the out-of-plane distortion. Altering the
residual stress in the at least one plate 200 may be accomplished
in a number of ways. In one particular embodiment, the step of
altering the residual stress in the at least one plate 200 further
comprises the step of disrupting at least one compressive stress
pattern in the at least one plate 200 by inducing at least one area
of tensile stress.
In an illustrative embodiment, shown in FIG. 3., of the present
invention, two welding devices 310, 320, and two heat sources 410,
420; located on opposing sides of a stiffener 100, are used. A
first welding device 310 and a second welding device 320, located
on opposing sides of the stiffener 100, may concurrently weld the
opposing sides of the stiffener 100 to the plate 200. A first heat
source 410 may be located at a first predetermined lateral distance
414 from a stiffener 100 and a second heat source 420 may be
located at a second predetermined lateral distance 424 from the
stiffener 100, on the opposing side. The first 410 and second 420
heat sources act to introduce tensile residual stress thereby
altering the normal compressive residual stress profile.
As one with skill in the art can appreciate, the at least one
welding device 300 may be any fusion welding equipment. The at
least one welding device 300 would most likely be an arc welding
device in most widespread applications. Similarly, the at least one
heat source may include virtually any means for transferring heat
from an external source to the at least one plate 200. Such heating
sources may be in contact with the at least one plate 200 or may be
separated by a predetermined separation distance 414 from a surface
210 of the at least one plate. In one particular embodiment it has
been found that flame heaters have provided a cost effective
heating source of minimal complexity. Such a configuration requires
minimal effort to implement into existing stiffener 100 and plate
200 welding arrangements, as might be common in a shipyard. In this
embodiment, the flame heaters are held away from the at least one
plate 200 by the predetermined separation distance 430 and they may
be fueled by virtually any combustible, most commonly natural gas.
The flame from the flame heaters may extend across the
predetermined separation distance 430, thereby coming in contact
with the at least one plate 200, or the flame may be held away from
the surface 210 of the at least one plate 200. In additional
embodiments, the heat source may be supplied by resistance heating,
induction heating, heat pads, the direct application of electrical
current to the plate, or by various other methods of heat
application, as would be understood by one skilled in the art.
The at least one welding device 300 and the at least one heat
source 400 may move concurrently during the welding process,
thereby not requiring additional time to fabricate a stiffened
panel. This may be achieved through mounting them on a common
motion control system, or through the use of more advanced
automated motion control systems. The at least one heat source may
lead, follow, or align with the at least one welding device as
welding occurs.
As one with skill in the art can appreciate, once it is recognized
that transient thermal tensioning to minimize distortion should not
focus on minimizing residual stress, but rather on minimizing the
propensity to distort, a number of techniques may be used in
determining optimal locations to apply the at least one heat
source. One such method particularly well suited for this
determination is finite element analysis (FEA) modeling. FEA is a
technique for predicting the responses of structures and materials
to environmental factors such as force, heat, and vibration. The
process starts with the creation of a geometric model, which is
then divided into smaller shapes connected at specific nodal
points. In this manner, stress-strain relationships are more easily
approximated. Finally, the material behavior and boundary
conditions are applied to each element, and the analysis is
performed.
Traditional FEA of distortion in stiffened panels have shown a good
correlation with experimental results. As one with skill in the art
can appreciate, distortion in a panel sets in only after the panel
has cooled. The term panel herein refers to the combined
manufacture of at least one stiffener 100 welded to at least one
plate 200. During welding of the at least one stiffener 100 to the
at least one plate 200, the geometry of the panel remains within
small-deformation bounds. Thus, it has been recognized that the
buckling analysis and residual stress analysis can be decoupled.
Residual stresses can be determined on the unbuckled panel and then
the stresses can be used as loading for a buckling analysis. Prior
techniques then create a "welding load" in the form of a
contraction at the weld line computed from the residual stress and
apply this welding load to a 3D structural model of the panel for
further analysis. Buckling analysis is then performed on the loaded
model thereby yielding the propensity of the panel to buckle.
The methods to predict buckling include a series of steps that arc
performed in order, specifically: 1) a thermal model for
calculation of temperature histories throughout the plate 200, 2) a
method to predict the residual stress induced in the plate 200 by
the welding process and any auxiliary heat application, and 3) an
eigenvalue solver to determine the buckling modes. The thermal
analysis is performed by using any suitable finite element analysis
technique.
Upon the discovery that transient thermal tensioning to minimize
distortion should not focus on minimizing residual stress, but
rather on minimizing the propensity to distort, the present
inventors recognized a preferred method of analysis. The preferred
method of analysis is just one example of modeling transient
thermal tensioning to minimize distortion based upon minimizing the
propensity to buckle, and shall not be construed to be limiting on
the scope of the present invention. It has been determined that the
previously mentioned step of computing a welding load is
unnecessary. Due to the possibility of utilizing several heating
locations and patterns, computing the welding load is replaced by
transferring residual stresses from the stress analysis directly to
the buckling analysis. The optimum heater locations are then
determined by modeling several cases and selecting one with the
least propensity to buckle. This approach is applied to a
laboratory mock-up panel and to a full sized shipyard panel, with
the results explained below.
Residual Stress Prediction
The weld process is well approximated to be one way coupled
thermo-elastoplastic. Transient heat-transfer analyses simulate the
temperature history in the structure due to welding. The
temperature history is used as loading for a quasi-static
elastoplastic analysis. This approach has been shown to yield good
estimations of residual stresses.
Due to the computational expense of full three-dimensional (3D)
models, the weld process is often approximated in two dimensions
(2D). Such a 2D model consists of a cross section of the welded
structure normal to the welding direction. In a typical panel, the
cross section of the panel is uniform in the welding direction. The
welding parameters like heat input, etc. are also kept constant.
Any number of finite element analysis techniques may then be used
to compute the residual stress.
The experimental panel consists of a plate 200 of size 8.times.20
feet, and is 3/16.sup.th inch thick. Three tee-shaped stiffeners
100, each 4.times.4 inches are fillet welded to the plate 200 along
the 20 foot direction, similar to the configuration illustrated in
FIG. 4. The stiffeners 100 are placed 2 feet apart from the edges
and each other. The fillet welds are 3/16 inch leg size and are
continuous. The experimental weld process is flux cored arc
welding. Welding on both sides of the stiffener 100 is carried out
simultaneously with an electrode offset of 4 inches. Both the plate
200 and the stiffeners 100 are made of AH36 steel. Standard thermal
and mechanical material properties are used for the analysis.
The panel is modeled using 20-noded brick elements. The analysis of
the experimental panel is carried out in three passes. Each pass
involves the welding of a single stiffener 100 and application of
the two heat sources 410, 420. The results at the outlet of the
model are used as inlet conditions for the subsequent pass. The
analysis is one way coupled. Three thermal solutions representing
each pass are used to load the three subsequent elastoplastic
simulations.
In the experiment, the heat input of each weld was 15 KJ/inch. It
is applied in a Gaussian distribution over a double ellipsoid
volume. The heat sources were flame heaters having a length of 6
inches and they were aligned with the electrode. The heat input at
each heater is 11.2 KJ/inch. The heat input for the flame heater is
applied in a Gaussian distribution on a rectangular area. The
travel speed is 23 inch/minute. In different embodiments, the heat
input of the weld and of the heat sources may be variable, and in
particular, those embodiments using a plurality of heat sources may
operate with various heat sources operating at variable heat
input.
Buckling Analysis
The residual stresses at the outlet of the 3.sup.rd pass of
eulerian analysis are applied along the entire length of the
buckling model. The stresses are ramped to zero at the start and
end edges to simulate free stress conditions. The stress stiffness
matrix k.sub..sigma. is assembled directly from the stresses as
follows. ##EQU1##
Where the stress matrix s given by ##EQU2##
Matrix G operates on the nodal displacements u so that
The eigenvalue problem solved to obtain the mode shapes and
associated eigenvalues is
where k is the elastic stiffness, e are the eigenvectors and
.lambda. are the eigenvalues. The lowest eigenvalue indicates the
fraction of applied load sufficient to make the structure buckle.
Thus low eigenvalues indicate high buckling propensity. Eigenvalues
above unity indicate that buckling will not occur.
Boundary conditions are difficult to prescribe for the buckling
analysis. This is because the supporting conditions for the
structure depend on the buckled shape, which is not known a priori.
Thus, the panel is constrained by forcing all the stiffener lines
to lie in the same horizontal plane. Upon examination of the
experimental buckled panels, this assumption was found to be
valid.
Correlation with Experiments
Experimental panels were fabricated in the laboratory with varying
tensioning dimensions. Table 1 shows thermal tensioning parameters
for seven panels fabricated in the laboratory. The tensioning
dimensions, indicated by TD1 and TD2 in the table, indicate the
distance in inches of the auxiliary heater from the stiffener line,
and are analogous to the previously discussed predetermined lateral
distances 414, 424. Values of zero indicate no auxiliary heating
used at the location, e.g., Experiment 1 reflects a control
experiment where no thermal tensioning was used.
For example, Experiment 1, an untensioned plate with three welded
stiffeners, showed a low eigenvalue (0.21455) which correlated with
a large degree of out-of-plane distortion in the welded panel, as
seen in FIGS. 9, 11, 13, and 15. Contrariwise, Experiment 7, a
thermally tensioned plate wherein the thermal tensioning was
applied at the tensioning dimensions as seen in Table 1, showed a
much higher eigenvalue, which correlated with less out-of-plane
distortion, as seen in FIGS. 8, 10, 12, and 14.
TABLE 1 Thermal Tensioning Cases Pass 1 Pass 2 Pass 3 Experiment
TD1 TD2 TD1 TD2 TD1 TD2 Eigenvalue 1 0 0 0 0 0 0 0.21455 2 10 10 10
10 10 10 0.14721 3 20 20 20 20 20 20 0.42858 4 20 20 10 10 10 20
0.42277 5 20 20 10 20 10 20 0.40588 6 20 0 0 0 0 20 0.8801 7 21 0
44 44 0 21 0.8882
The experimental buckling mode shapes and eigenvalues compared well
with predicted mode shapes. Further, the magnitude of the
displacements compares well with the eigenvalues, as can be seen in
comparing the eigenvalue in the experimental plate made without
transient thermal tensioning, that is, comparing Experiment 1 of
Table 1 with the out-of-plane distortions seen in FIGS. 9, 11, 13
and 15. Similarly, comparing the eigenvalue in the plate made with
optimal transient thermal tensioning, that is, Experiment 7 of
Table 1, shows a reduced level of out-of-plane distortion seen in
FIGS. 8, 10, 12, and 14. In general, the larger the eigenvalue, the
less the out-of-plane distortion, with eigenvalues greater than
unity indicating no propensity to buckle.
Minimizing Buckling Distortion
The experimental correlation indicates that a given auxiliary heat
pattern can be evaluated adequately. In absence of a rigorous
formulation for the sensitivity of the eigenvalues to the auxiliary
heat pattern, the approach taken was to exhaust the design space,
and choose the best auxiliary heat pattern. Due to the large number
of trials, residual stress analyses were not conducted for each
trial. The residual stress due to a single auxiliary heater and a
single weld pass were merged to obtain the appropriate overall
stress profile. The auxiliary heaters were far enough from the weld
line for the superimposition to be valid. However, residual
stresses due to adjacent heaters have a small error due to
merging.
Due to the stiffeners on the plate, the buckling modes involve
out-of-plane edge waviness and waviness between stiffeners. The
stress-stiffness of these areas may be increased by introducing
residual tensions in these areas, in this particular case.
Similarly, in this particular experiment, it was found that
applying auxiliary heat between the stiffeners reduces the
eigenvalues. Thus, only the effect of heaters on the edges was
investigated.
Transient Thermal Tensioning in an Experimental Plate with Three
Stiffeners
The longitudinal stress profile for a representative example of an
experimental plate 200 welded with three longitudinal stiffeners
100 appears in FIG. 16. The top graph of FIG. 16, indicated as Pass
One, shows the longitudinal stress profile of the experimental
plate 200 after the welding of a single stiffener 100, as shown in
FIG. 5, according to the transient thermal tensioning method of the
instant invention. The single stiffener 100 has been welded
longitudinally at approximately 24 inches from a first edge of the
plate 200, and the welding of the single stiffener 100 has created
an area of residual tensile stress of that rises to a level of
approximately 750 MPa. Just lateral to the weld lines, on both
sides of the single stiffener 100, are areas of residual
compressive stress falling to levels of approximately -200 MPa.
Transient thermal tensioning according to the method of the instant
invention, applied approximately 20 inches lateral to the single
stiffener 100, have induced additional areas of residual tensile
stress, rising to levels of approximately 375 MPa.
The middle graph of FIG. 16, indicated as Pass Two, shows the
longitudinal stress profile of the experimental plate 200 after the
welding of a second stiffener 100, as shown in FIG. 6, according to
the transient thermal tensioning method of the instant invention.
The second stiffener 100 has been welded longitudinally at
approximately 48 inches from the first edge of the plate 200, and
the welding of the single stiffener 100 has created an area of
residual tensile stress of that rises to a level of approximately
750 MPa. Just lateral to the weld lines, on both sides of the
second stiffener 100, are areas of residual compressive stress
falling to levels of between approximately -150 and -200 MPa.
Transient thermal tensioning according to the method of the instant
invention, applied approximately 38 and 70 inches from the first
edge of the plate 200, and therefore lateral to the second
stiffener 100, have induced additional areas of residual tensile
stress, rising to levels of approximately 375 MPa.
The bottom graph of FIG. 16, indicated as Pass Three, shows the
longitudinal stress profile of the experimental plate 200 after the
welding of a third stiffener 100, as shown in FIG. 7, according to
the transient thermal tensioning method of the instant invention.
The third stiffener 100 has been welded longitudinally at
approximately 72 inches from the first edge of the plate, and the
welding of the single stiffener 100 has created an area of residual
tensile stress of that rises to a level of approximately 750 MPa.
Just lateral to the weld lines, on both sides of the second
stiffener 100, are areas of residual compressive stress falling to
levels of between approximately -150 and -200 MPa. Transient
thermal tensioning according to the method of the instant
invention, applied approximately 62 and 92 inches from the first
edge of the plate, and therefore lateral to the third stiffener
100, have induced additional areas of residual tensile stress,
rising to levels of approximately 375 MPa
Experimental Conclusions
Experimentation confirmed that the method of transient thermal
tensioning of the present invention greatly minimizes distortion in
welded stiffened plates 200. Further, finite element modeling of
the process has been demonstrated and used to determine the optimum
locations of the at least one heat source 400 to minimize
distortion. Therefore, the method of the present invention
successfully reduces the propensity of a stiffened plate 200 to
buckle by disrupting the large expanses of compression that exist
between stiffeners 100, without making any attempt to reduce the
peak stress that is the focus of prior distortion reducing
attempts.
Numerous alterations, modifications, and variations of the
preferred embodiments disclosed herein will be apparent to those
skilled in the art and they are all anticipated and contemplated to
be within the spirit and scope of the instant invention. For
example, although specific embodiments have been described in
detail, those with skill in the art will understand that the
preceding embodiments and variations can be modified to incorporate
various types of substitute, and/or additional or alternative
materials, relative arrangement of elements, and dimensional
configurations. Accordingly, even though only a few variations of
the present invention are described herein, it is to be understood
that the practice of such additional modifications and variations
and the equivalents thereof, are within the spirit and scope of the
invention as defined in the following claims.
The corresponding structures, materials, acts, and equivalents of
all means or step plus function elements in the claims below are
intended to include any structure, material, or acts for performing
the functions in combination with other claimed elements as
specifically claimed.
* * * * *